Bees are capable of fanning their wings to cause a draught, or movement of air, for cooling the hive or evaporating moisture from nectar.
A beehive with it's regular spaced rows of frames is convenient for the beekeeper, but may not be so suited to the bees needs as regards ease of ventilation.
If we can take more notice of the bees requirements then we will reduce the need for some of the fanning bees, thus releasing them for other duties and improving overall productivity.
Ventilation during clustering... There are two schools of thought on this matter one is to provide for a through draught of air and the other is to close off all forms of ventilation at the top of a hive (as indeed the bees do in the wild) and insulate the top cover.
I am at risk of upsetting many American beekeepers that "swear by" upper entrances which are deeply rooted in American bee management.
Wedmore wrote a book "The Ventilation of Beehives". He made many errors in his book simply because he had decided on the answer before doing the experimental work. He proposed the adoption, and deliberate inclusion of top ventilation by match sticks being placed under crown boards, leaving feed holes open, I am not sure whether he mentioned upper entrances, but they fall into the same category.
He obviously did not do much experimentation, or observe the way the bees do it, otherwise he would have found that the majority of bees found in hollow trees have a single entrance, of around 100 square mm, well below the halfway point of the nest.
If you give bees a mesh panel over an upper vent they will propolise it shut when they do not want such ventilation and will unblock the holes at times when they do want it. (I would like to understand how they "know" where a blocked up hole is so that they can unblock it, but that is a side issue.) I believe that bees propolise mesh and perforated metal because they don't recognise it, so are concerned about defence. I think this is also why a bait hive that has an OMF is unlikely to attract a swarm. R.P.
If you have a closed box with a heat source in it, a convection current will be set up. If that box is roughly cubic in form and the heat source is in the centre the air will rise up the centre and spread across the inside of the top like a mushroom, then fall down all outer surfaces before combining again at the bottom to replace the air that was displaced by the original rise. This is a dynamic process, the speed of which is governed by the energy input (heat from the bees), the density and viscosity of the gas, the "U" value of the box walls and the surface roughness of those walls.
Now our beehive is a little different... let us consider one with a full width entrance slot at the bottom of the "front", a thin plywood cover board (inner Cover) and all upper holes propolised shut with a wintering cluster of bees in the centre of the box. The whole hive probably has a telescoping roof, but this does not impinge on the conditions we have postulated as the outer surface of the thin plywood is at ambient temperature.
Our convection current will be set up as before, but most of the heat will be lost through the thin plywood top causing much condensation of the exhaled water vapour... This causes a circular wet patch that is wettest in the centre (sometimes the periphery) and drips will fall directly on the clustering bees, causing disturbance and a greater saturation of moisture in the rising air. The convection current continues around the circuit and a small portion of it will be "exchanged" with fresh air at the entrance.
Now consider the same circumstances, but with an insulated, sealed roof containing at least 50 mm of expanded polystyrene foam... The moist air will rise, but after a very minimal amount of condensation the temperature on the underside of the top will rise to close on the temperature of the rising stream of air and thus will not lose moisture due to condensation. Some condensation will occur over the large area of the sidewalls, but much will stay in suspension in the warmer air and so the air that is exchanged at the entrance contains a higher percentage of this suspended water.
To take this further we can consider replacing our solid floor with an open mesh one... As in the last case our air reaches the floor carrying much more water vapour and so a greater exchange of wet and dry air occurs at our screen barrier, than would have been the case with our single entrance slot.
In a hollow tree a different mechanism is at work... By the very nature of hollow trees the cavity is roughly cylindrical, but the upper and lower ends of it are rough and fibrous. Our colony produces a rising stream of moist air in exactly the same manner as our colony in the beehive, but this rising stream condenses on a fibrous end grain "roof" and is absorbed by capillary action, then widely disseminated throughout the structure of the tree. Any moisture that is still left in suspension by the time the airflow reaches the floor will encounter a rough "forest" of torn fibres that are cold and damp and yet more of the moisture will condense out and wick down the fibres adding water to the natural recycling/composting action that takes place under the nest. This leaves little extra water vapour in the air by the time it reaches the entrance where a portion of what is left will be exchanged for "fresh".
Deliberate ventilation at the top of the hive will cause some of the moist air to be lost, giving the appearance of improving the situation, but it will destroy the natural circulation of air within the hive, replacing it with a chimney effect.
I have been using solid, insulated roofs for 19 or 20 years now (without a separate coverboard). I have used them with solid floors, mesh floors, small entrances, large entrances, in damp wooded areas and on a city rooftop 130 feet up... I have not seen any condensation during this time.
The above "sets the scene" for the effect that this has on evaporation of nectar... As the circulation in winter is driven by heat from the cluster so summer circulation is driven by the difference in temperature of the brood nest and the rest of the hive, but this is not enough for the bees to shift the moisture caused in this artificial environment that we call a bee hive. The bees are not wasteful and so have discovered that the easiest and most efficient way to shift this moisture is to increase the natural circulation by fanning. The relative humidity of air within this circulation governs the amount of additional moisture the air can carry as vapour.
Convection and flow within the individual cell. As the cell is deeper than it is wide there is a space near it's bottom that has relatively stationary saturated air. This stagnant air is only moved by a bee entering the cell and displacing it. Near the mouth of the cell small circulating currents will be set up that are caused by the airflow past the cell concerned and this will homogenise the moisture levels of the air in the cell mouth and the main airflow.
The local temperature variation in bee hives. Bees have a remarkable ability to regulate temperature very accurately. But this does not mean that the temperature within a beehive is the same throughout. Bees will locally regulate temperature and humidity according to what is going on in the part of the hive concerned. For instance queen cells within a hive will receive "special" temperature and humidity control. I suspect that under emergency conditions that the bees may well raise the "normal" queen incubation temperature to hasten emergence, regardless of what other effects that this may produce.